Power system inertia adjusting method and system based on power grid information

Abstract

The invention discloses a power system inertia adjustment method based on power grid information, and relates to the field of power system automation, and the method comprises the steps: constructing a regional inertia parameter identification model, inputting the preprocessed data into the regional inertia parameter identification model, and carrying out the first regional inertia identification of a power system; the preprocessed data are fused and input into a regional inertia parameter identification model, and secondary regional inertia identification is carried out on the power system; and dynamically adjusting the inertia of each region according to the difference between the inertia of each region identified twice, and optimizing inertia distribution in real time through closed-loop control. Through a series of steps of data acquisition and preprocessing, intelligent partitioning and adaptive inertia distribution, multi-source data fusion inertia identification technology, dynamic adjustment and closed-loop control and the like, precise management and dynamic adjustment of the inertia of the power system are realized, and the operation stability and the capability of adapting to complex working conditions of the power system are improved. Therefore, the reliability and the economical efficiency of the whole power system are improved.

Description

Technical Field

[0001] The present invention relates to the field of power system automation, and in particular to a power system inertia adjustment method and system based on power grid information. Background Art

[0002] With the rapid development of power systems, especially the large-scale access of renewable energy sources such as wind power and solar energy, the inertia characteristics of power systems have changed significantly. Traditionally, the inertia of power systems is mainly provided by synchronous generators, while new energy power generation equipment, due to its inherent characteristics, usually does not have sufficient inertia to maintain the stable operation of the system. Therefore, how to effectively manage and regulate the inertia of power systems has become an important research topic. In recent years, the application of wide-area measurement systems has made it possible to dynamically monitor power systems. Synchronous phasor measurement units can collect real-time operating data of power systems, including key indicators such as node frequency, generator status, and active power, providing an important basis for state estimation and control of power systems. However, due to the high complexity and uncertainty of the power system itself, how to extract useful information from massive data to achieve accurate system state estimation and control is still a problem that needs to be solved.

[0003] First, in the data processing link, due to the presence of various types of noise interference in the power system, such as impulse noise and Gaussian noise, traditional data preprocessing methods are difficult to effectively remove these noises, resulting in inaccurate subsequent data analysis results. Secondly, in the process of inertia parameter identification, traditional identification methods often rely on a single data source and lack comprehensive consideration of multi-source heterogeneous data, which limits the accuracy and reliability of the identification results. The above problems limit the application effect of the prior art in complex power systems. The present invention aims to improve the accuracy and robustness of inertia parameter identification by introducing advanced data preprocessing technology and multi-source data fusion methods, thereby achieving effective management and dynamic regulation of power system inertia. Summary of the invention

[0004] In order to solve the problem of insufficient inertia parameter identification accuracy in traditional methods, the present invention provides the following technical solutions: In a first aspect, the present invention provides an inertia regulation method for an electric power system based on power grid information, comprising: real-time collection of electric power data, and preprocessing of the electric power data; partitioning of node frequencies through multiple clustering analyses; constructing a regional inertia parameter identification model, inputting the preprocessed data into the regional inertia parameter identification model, and performing a first regional inertia identification on the electric power system; fusing the preprocessed data and inputting the data into the regional inertia parameter identification model, and performing a second regional inertia identification on the electric power system; dynamically adjusting the inertia of each region according to the difference between the inertias of each region identified twice, and optimizing the inertia distribution in real time through closed-loop control.

[0005] As a preferred solution of the power system inertia adjustment method based on power grid information of the present invention, the real-time collection of power data and preprocessing of the collected data are specifically performed as follows: Real-time acquisition of node frequency, synchronous generator status and active power of new energy stations through wide-area measurement, synchronous phasor measurement units and smart meters; Median filtering and mean filtering are used to remove impulse noise and Gaussian noise.

[0006] As a preferred solution of the power system inertia adjustment method based on power grid information of the present invention, the node frequency is partitioned by multiple clustering analyses, and the specific steps are as follows: The collected node frequencies are standardized, and the standardized node frequency sequences are clustered multiple times using the k-shape algorithm; Multiple random selections The shapes are used as the initial centroids, and each data point is assigned to the cluster represented by the most similar centroid. The centroid of each cluster is recalculated until the centroid no longer changes significantly or the maximum number of iterations is reached. The results of each clustering are recorded, including the members and centroids of each cluster. Evaluation indicators are used to measure the clustering quality under different cluster numbers. Gradually increase the number of clusters , record the results and evaluation indicators of each clustering, and combine the elbow method and Silhouette method to determine the optimal number of clusters ; The optimal number of clusters Refers to the optimal number of node frequency partitions.

[0007] As a preferred solution of the power system inertia adjustment method based on power grid information of the present invention, wherein: the regional inertia parameter identification model is constructed, the pre-processed data is input into the regional inertia parameter identification model, and the first regional inertia identification of the power system is performed, and the specific steps are as follows: Based on the rotor swing equation, the regional inertia parameter identification model is constructed, and the expression is: ; in, It is a region The equivalent inertia of It is a region The active power change, is the rated frequency, It is a region Frequency deviation; The adaptive volumetric Kalman filter algorithm is used for parameter identification. By establishing a state space model and initializing the state, the state estimation and covariance matrix are continuously adjusted using the prediction and update steps, the global equivalent inertia is gradually refined, and the inertia of each region is generated. .

[0008] As a preferred solution of the power system inertia adjustment method based on power grid information of the present invention, wherein: the fusion of pre-processed data and input into the regional inertia parameter identification model to perform a second regional inertia identification on the power system, the specific steps are as follows: The preprocessed node frequency, synchronous generator state and active power of the new energy station are standardized respectively, and the convolutional neural network is used for training. The expression is: ; in, is the normalized node frequency, is the normalized active power of the synchronous generator, is the standardized active power of new energy stations, It is Convolutional neural network parameters for each data source; Use federated learning technology for joint training, the expression is: ; in, are the parameters after joint training, is the total amount of data from all data sources, It is The convolutional neural network parameters of the data source, It is The amount of data from each data source, is the number of data sources; Using the federated learning model for data fusion, the expression is: ; in, is the fused data, is the normalized node frequency, is the normalized active power of the synchronous generator, is the standardized active power of new energy stations; The fused data is used to calculate the active power change and frequency deviation, and the fused regional inertia is calculated through the rotor swing equation. The expression is: ; in, is the regional inertia after fusion, is the rated frequency.

[0009] As a preferred solution of the power system inertia adjustment method based on power grid information of the present invention, the inertia of each area is dynamically adjusted according to the difference between the inertias of each area identified twice, and the specific steps are as follows: Dynamic adjustment strategy based on adaptive PID controller; The error signal is defined based on the deviation between the expected frequency and the actual frequency to measure the deviation between the current state and the target state. The expression is: ; in, is the error signal, is the expected frequency, is the actual frequency; Inertia of each area Initialize the PID parameters and use the initialized PID parameters to calculate the control output. The expression is: ; in, To control the output, is the proportionality coefficient, is the integration coefficient, is the differential coefficient; Through the fusion of inertia Adjust the PID controller parameters and use the adjusted PID parameters to calculate the adjusted control output .

[0010] The specific steps of optimizing inertia distribution in real time through closed-loop control are as follows: The control output is obtained by calculating twice and , respectively applied to the power system to adjust the active power output of the power system, and the node frequency data and error signal are calculated again through the adjusted active power; Continuously adjust PID parameters and inertia through closed-loop control, set the error signal threshold, stop adjustment when the error signal reaches this threshold, set the maximum number of iterations, if the error signal has not reached the threshold, reach the number of iterations, stop adjustment; The control gain is defined based on the proportional relationship between the error signal and the inertia change, which is used to quantify the influence of the control output on the behavior. The expression is: ; in, To control the gain, is the change in inertia of two regional inertia identifications; Calculate the adjusted area inertia, the expression is: ; in, is the adjusted regional inertia, is the adjustment factor.

[0011] In a second aspect, the present invention provides an inertia regulation system for an electric power system based on power grid information, including a data acquisition module for real-time acquisition of electric power data and preprocessing of the electric power data; a frequency partitioning module for partitioning node frequencies through multiple clustering analyses; an inertia parameter identification module for constructing a regional inertia parameter identification model, inputting the preprocessed data into the regional inertia parameter identification model, and performing a first regional inertia identification on the electric power system; an inertia adjustment module for fusing the preprocessed data and inputting the data into the regional inertia parameter identification model, and performing a second regional inertia identification on the electric power system; and a closed-loop control module for dynamically adjusting the inertia of each region according to the difference between the inertias of each region identified twice, and optimizing the inertia distribution in real time through closed-loop control.

[0012] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the power system inertia regulation method based on power grid information as described in the first aspect of the present invention is implemented.

[0013] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the power system inertia regulation method based on power grid information as described in the first aspect of the present invention is implemented.

[0014] The beneficial effects of the present invention are as follows: through a series of steps such as data acquisition and preprocessing, intelligent partitioning and adaptive inertia allocation, inertia identification technology of multi-source data fusion, dynamic adjustment and closed-loop control, accurate management and dynamic adjustment of the inertia of the power system are achieved. In particular, in the data preprocessing stage, median filtering and mean filtering are used to remove impulse noise and Gaussian noise, which further improves the data quality and provides a solid foundation for subsequent intelligent partitioning and adaptive inertia allocation. Finally, the present invention improves the operating stability of the power system and the ability to adapt to complex working conditions, thereby improving the reliability and economy of the entire power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0016] Figure 1This is a flow chart of the method for adjusting the inertia of a power system based on the power grid information in Example 1.

[0017] Figure 2 This is a flow chart of the power system inertia regulation system of the power grid information in Example 1. DETAILED DESCRIPTION

[0018] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0019] Example 1 Reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a method for adjusting the inertia of a power system based on power grid information, comprising the following steps: S1: collect power data in real time and pre-process the power data; Furthermore, wide-area measurement, synchronous phasor measurement units, and smart meters collect node frequency, synchronous generator status, and active power of new energy stations in real time; wide-area measurement usually consists of multiple synchronous phasor measurement units, which can provide synchronized, high-precision phasor measurements. PMU devices are installed at key nodes of the network to collect instantaneous values ​​of voltage and current, and then synchronize these measurements using global positioning timestamps; Median filtering and mean filtering are used to remove impulse noise and Gaussian noise; It should be noted that the median filter replaces the original data value at the center of the window by sliding a window of a selected size in the data stream and sorting the data in the window and taking the median. This method can effectively remove impulse noise. Even if there are extreme values ​​in the window, the median can resist the influence of such abnormal values, thereby retaining the basic characteristics of the signal without interference. Mean filtering replaces the original data value at the center of the window by averaging the data points within the window. This method is suitable for smoothing signals and removing Gaussian distributed random noise, because the average value can reduce the impact of the extreme value of a single data point on the overall signal, making the signal smoother and retaining the basic trend and pattern of the signal.

[0020] S2: Partitioning node frequencies through multiple clustering analyses; Furthermore, the collected node frequencies are standardized, and the standardized node frequency series are clustered multiple times using the k-shape algorithm. The mean and standard deviation of each frequency time series are calculated, and the data is transformed using standardization. The expression is: ; in, is the frequency time series data, is the average value, is the standard deviation; Multiple random selections The shapes are used as the initial centroids, and each data point is assigned to the cluster represented by the most similar centroid. The centroid of each cluster is recalculated until the centroid no longer changes significantly or the maximum number of iterations is reached. The results of each clustering are recorded, including the members and centroids of each cluster. Evaluation indicators are used to measure the clustering quality under different cluster numbers. Based on historical data experience, k time series are randomly selected as the initial centroids. Gradually increase the number of clusters , record the results and evaluation indicators of each clustering, and combine the elbow method and Silhouette method to determine the optimal number of clusters ; It should be noted that the Silhouette method measures the closeness of a sample to its cluster and the degree of separation from other clusters, and the elbow method observes the changes in the clustering cost function corresponding to different k values; The optimal number of clusters Refers to the optimal number of node frequency partitions.

[0021] S3: construct a regional inertia parameter identification model, input the preprocessed data into the regional inertia parameter identification model, and perform the first regional inertia identification of the power system; Furthermore, based on the rotor swing equation, a regional inertia parameter identification model is constructed, and the expression is: ; in, It is a region The equivalent inertia of It is a region The active power change, is the rated frequency, It is a region In the power system, the rotor swing equation describes the relationship between the change of the synchronous generator rotor angular velocity and the active power imbalance in the power system, and the expression is: ; in, It is a region Changes in the angular velocity of the internal generator; The adaptive volumetric Kalman filter algorithm is used for parameter identification. By establishing a state space model and initializing the state, the state estimation and covariance matrix are continuously adjusted using the prediction and update steps, the global equivalent inertia is gradually refined, and the inertia of each region is generated. .

[0022] It should be noted that by defining the state equation and observation equation, selecting a reasonable initial state and covariance matrix, and using the Kalman filter algorithm to continuously adjust the state estimate and covariance matrix, the process noise and observation noise are adjusted to improve the robustness and accuracy of the filter.

[0023] S4: Fusing the preprocessed data and inputting them into the regional inertia parameter identification model, a second regional inertia identification is performed on the power system; Furthermore, the preprocessed node frequency, synchronous generator state and active power of the new energy station are standardized and trained using a convolutional neural network. The expression is: ; in, is the normalized node frequency, is the normalized active power of the synchronous generator, is the standardized active power of new energy stations, It is The convolutional neural network parameters of the data source are calculated, and the CNN architecture of the convolutional neural network for processing time series data is constructed. The local features are extracted through the convolution layer, the feature dimension is reduced through the pooling layer, and the fully connected layer is used for classification. The standardized data is input into the CNN for training, and the back propagation algorithm is used to optimize the network parameters. , which enables the network to learn the intrinsic patterns and correlations in the data. CNN can automatically learn the local features and long-range dependencies in the data, which is very suitable for processing time series data. Through training, CNN can capture important patterns in frequency and power changes, providing a basis for subsequent fusion and analysis; Use federated learning technology for joint training, the expression is: ; in, are the parameters after joint training, is the total amount of data from all data sources, It is The convolutional neural network parameters of the data source, It is The amount of data from each data source, The number of data sources. Federated learning allows joint training using data from different data sources while protecting privacy. This allows joint optimization of the model through parameter updates without sharing actual data. Using the federated learning model for data fusion, the expression is: ; in, is the fused data, is the normalized node frequency, is the normalized active power of the synchronous generator, For the standardized active power of renewable energy stations, a comprehensive and more representative feature representation can be obtained by fusing multi-source data through a jointly trained model, which is helpful for subsequent analysis and decision-making, especially when multiple factors need to be considered comprehensively; The fused data is used to calculate the active power change and frequency deviation, and the fused regional inertia is calculated through the rotor swing equation. The expression is: ; in, is the regional inertia after fusion, For the rated frequency, the active power change and frequency deviation are calculated by the fused data. Combined with the rotor swing equation, a more accurate and comprehensive inertia can be obtained, which is helpful to dynamically adjust the inertia of each area and optimize the operation of the power system.

[0024] S5: According to the difference between the inertia of each area identified twice, the inertia of each area is dynamically adjusted, and the inertia distribution is optimized in real time through closed-loop control; Based on the dynamic adjustment strategy of adaptive PID controller, the adaptive PID controller can automatically adjust the controller parameters according to the real-time state of the power system and external disturbances, thereby improving the stability and response speed of the control system; The error signal is defined based on the deviation between the expected frequency and the actual frequency to measure the deviation between the current state and the target state. The expression is: ; in, is the error signal, is the expected frequency, is the actual frequency; The error signal calculates the difference between the actual frequency and the expected frequency, and intuitively understands the deviation between the current state of the power system and the target state. The error signal can be used as the input of the controller to adjust the control output so that the power system can reach the desired frequency state as soon as possible. In the dynamic adjustment strategy, the error signal provides real-time feedback to help the controller dynamically adjust the control parameters according to the current deviation to achieve better control performance; Inertia of each area Initialize the PID parameters and use the initialized PID parameters to calculate the control output. The expression is: ; in, To control the output, is the proportionality coefficient, is the integration coefficient, is the differential coefficient; Using adaptive algorithms to adjust PID control parameters , , , the expression is: ; ; ; in, is the learning rate of the proportional coefficient, The learning rate of the integration coefficient, is the learning rate of the differential coefficient; By the fused regional inertia Adjust the PID controller parameters and use the adjusted PID parameters to calculate the adjusted control output , the expression is: ; in, To adjust the control output; The control output is obtained by calculating twice and , respectively applied to the power system to adjust the active power output of the power system. Through the adjusted active power, the node frequency data and error signal are calculated again to adjust the active power of the generator. The expression is: ; in, is the base state active power output, is the adjusted active power; Through closed-loop control, PID parameters and inertia are continuously adjusted, and the error signal threshold is set. When the error signal reaches this threshold, the adjustment is stopped. The maximum number of iterations is set. If the error signal has not reached the threshold, the adjustment is stopped when the number of iterations is reached. , check whether the error signal is less than or equal to the set error signal threshold; when ≤ When , stop adjusting, indicating that it is close enough to the expected frequency; when > Continue to adjust; The control gain is defined based on the proportional relationship between the error signal and the inertia change, which is used to quantify the influence of the control output on the behavior. The expression is: ; in, To control the gain, is the change in inertia of two regional inertia identifications; The regional inertia change is calculated using the fused regional inertia and the first identified regional inertia. The expression is: ; Control gain It provides a measure to quantify the degree of influence of the control output on the behavior. Through this gain, the influence of the control output on the frequency change can be better understood; Calculate the adjusted area inertia, the expression is: ; in, is the adjusted regional inertia, The adjustment factor is used to control the magnitude of the adjustment. A smaller value means a gentler adjustment, while a larger value means a larger adjustment. Choosing a suitable value can balance the speed and stability of the adjustment. It should be noted that the object of adjustment is the inertia of each area, and what is calculated is the difference in inertia of each area after two identifications.

[0025] This embodiment also provides an electric power system inertia regulation system based on power grid information, including: a data acquisition module, which collects electric power data in real time and pre-processes the electric power data; Frequency partitioning module, which partitions node frequencies through multiple clustering analyses; Inertia parameter identification module, builds a regional inertia parameter identification model, inputs the preprocessed data into the regional inertia parameter identification model, and performs the first regional inertia identification of the power system; The inertia adjustment module integrates the pre-processed data and inputs it into the regional inertia parameter identification model to perform the second regional inertia identification of the power system; The closed-loop control module dynamically adjusts the inertia of each area according to the difference between the inertia of each area identified twice, and optimizes the inertia distribution in real time through closed-loop control.

[0026] This embodiment also provides a computer device, which is applicable to the case of a power system inertia adjustment method based on power grid information, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to implement the power system inertia adjustment method based on power grid information as proposed in the above embodiment.

[0027] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0028] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for adjusting the inertia of the power system based on the power grid information proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0029] In summary, the present invention realizes the precise management and dynamic adjustment of the inertia of the power system through a series of steps including data acquisition and preprocessing, intelligent partitioning and adaptive inertia allocation, inertia identification technology of multi-source data fusion, dynamic adjustment and closed-loop control. In particular, median filtering and mean filtering are used to remove impulse noise and Gaussian noise in the data preprocessing stage, which further improves the data quality and provides a solid foundation for the subsequent intelligent partitioning and adaptive inertia allocation. Finally, the present invention improves the operating stability of the power system and its ability to adapt to complex working conditions, thereby improving the reliability and economy of the entire power system.

[0030] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the present invention, although the present invention has been described in detail with reference to the preferred embodiments.

Claims

1. A method for adjusting inertia of a power system based on power grid information, characterized in that: include: Collect power data in real time and pre-process the power data; Node frequencies were partitioned through multiple cluster analyses; Construct a regional inertia parameter identification model, input the preprocessed data into the regional inertia parameter identification model, and perform the first regional inertia identification of the power system; The pre-processed data is integrated and input into the regional inertia parameter identification model to perform the second regional inertia identification of the power system; According to the difference between the inertia of each area identified twice, the inertia of each area is dynamically adjusted, and the inertia distribution is optimized in real time through closed-loop control.

2. The method for adjusting the inertia of a power system based on power grid information according to claim 1, characterized in that: The real-time collection of power data and preprocessing of the collected data are carried out in the following specific steps: Real-time acquisition of node frequency, synchronous generator status and active power of new energy stations through wide-area measurement, synchronous phasor measurement units and smart meters; Median filtering and mean filtering are used to remove impulse noise and Gaussian noise.

3. The method for adjusting the inertia of a power system based on power grid information according to claim 2, characterized in that: The node frequencies are partitioned by multiple clustering analyses, and the specific steps are as follows: The collected node frequencies are standardized, and the standardized node frequency sequences are clustered multiple times using the k-shape algorithm; Multiple random selections The shapes are used as the initial centroids, and each data point is assigned to the cluster represented by the most similar centroid. The centroid of each cluster is recalculated until the centroid no longer changes significantly or the maximum number of iterations is reached. The results of each clustering are recorded, including the members and centroids of each cluster. Evaluation indicators are used to measure the clustering quality under different cluster numbers. Gradually increase the number of clusters , record the results and evaluation indicators of each clustering, and combine the elbow method and Silhouette method to determine the optimal number of clusters ; The optimal number of clusters Refers to the optimal number of node frequency partitions.

4. The method for adjusting the inertia of a power system based on power grid information according to claim 3, characterized in that: The regional inertia parameter identification model is constructed, the pre-processed data is input into the regional inertia parameter identification model, and the first regional inertia identification of the power system is performed. The specific steps are as follows: Based on the rotor swing equation, the regional inertia parameter identification model is constructed, and the expression is: ； in, It is a region The equivalent inertia of It is a region The active power change, is the rated frequency, It is a region Frequency deviation; The adaptive volumetric Kalman filter algorithm is used for parameter identification. By establishing a state space model and initializing the state, the state estimation and covariance matrix are continuously adjusted using the prediction and update steps, the global equivalent inertia is gradually refined, and the inertia of each region is generated. .

5. The method for adjusting the inertia of a power system based on power grid information according to claim 4, characterized in that: The fusion preprocessed data is input into the regional inertia parameter identification model to perform a second regional inertia identification on the power system. The specific steps are as follows: The preprocessed node frequency, synchronous generator state and active power of the new energy station are standardized respectively, and the convolutional neural network is used for training. The expression is: ； in, is the normalized node frequency, is the normalized active power of the synchronous generator, is the standardized active power of new energy stations, It is Convolutional neural network parameters for each data source; Use federated learning technology for joint training, the expression is: ； in, are the parameters after joint training, is the total amount of data from all data sources, It is The convolutional neural network parameters of the data source, It is The amount of data from each data source, is the number of data sources; Using the federated learning model for data fusion, the expression is: ； in, is the fused data, is the normalized node frequency, is the normalized active power of the synchronous generator, is the standardized active power of new energy stations; The fused data is used to calculate the active power change and frequency deviation, and the fused regional inertia is calculated through the rotor swing equation. The expression is: ； in, is the regional inertia after fusion, is the rated frequency.

6. The method for adjusting the inertia of a power system based on power grid information according to claim 5, characterized in that: The inertia of each region is dynamically adjusted according to the difference between the inertia of each region identified twice. The specific steps are as follows: Dynamic adjustment strategy based on adaptive PID controller; The error signal is defined based on the deviation between the expected frequency and the actual frequency to measure the deviation between the current state and the target state. The expression is: ； in, is the error signal, is the expected frequency, is the actual frequency; Inertia of each area Initialize the PID parameters and use the initialized PID parameters to calculate the control output. The expression is: ； in, To control the output, is the proportionality coefficient, is the integration coefficient, is the differential coefficient; By the fused regional inertia Adjust the PID controller parameters and use the adjusted PID parameters to calculate the adjusted control output .

7. The method for adjusting the inertia of a power system based on power grid information according to claim 6, characterized in that: The specific steps of optimizing inertia distribution in real time through closed-loop control are as follows: The control output is obtained by calculating twice and , respectively applied to the power system to adjust the active power output of the power system, and the node frequency data and error signal are calculated again through the adjusted active power; Continuously adjust PID parameters and inertia through closed-loop control, set the error signal threshold, stop adjustment when the error signal reaches this threshold, set the maximum number of iterations, if the error signal has not reached the threshold, reach the number of iterations, stop adjustment; The control gain is defined based on the proportional relationship between the error signal and the inertia change, which is used to quantify the influence of the control output on the behavior. The expression is: ； in, To control the gain, is the change in inertia of two regional inertia identifications; Calculate the adjusted area inertia, the expression is: ； in, is the adjusted regional inertia, is the adjustment factor.

8. An inertia adjustment system for a power system based on power grid information, based on the inertia adjustment method for a power system based on power grid information according to any one of claims 1 to 7, characterized in that: It includes data acquisition module, frequency partition module, inertia parameter identification module, inertia adjustment module and closed-loop control module; Data acquisition module, which collects power data in real time and pre-processes the power data; Frequency partitioning module, which partitions node frequencies through multiple clustering analyses; Inertia parameter identification module, builds a regional inertia parameter identification model, inputs the preprocessed data into the regional inertia parameter identification model, and performs the first regional inertia identification of the power system; The inertia adjustment module integrates the pre-processed data and inputs it into the regional inertia parameter identification model to perform the second regional inertia identification of the power system; The closed-loop control module dynamically adjusts the inertia of each area according to the difference between the inertia of each area identified twice, and optimizes the inertia distribution in real time through closed-loop control.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the power system inertia adjustment method based on power grid information as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the power system inertia adjustment method based on power grid information as described in any one of claims 1 to 7 are implemented.

Images